JustVugg 6e7aa6f92e Guard against running the tiny oracle against a real model (#76) ...

The default ./glm mode validates against ref_glm.json, which is generated
from glm_tiny (a random tiny model). Pointed at the 744B GLM-5.2 it feeds
the model tiny-vocab prompt tokens and compares against tiny references —
0/20 and a degenerate loop, guaranteed on EVERY platform (x86/ARM/Metal),
which reads as an engine bug (it isn't). Detect the mismatch (real vocab +
tiny-range oracle ids) and print the correct commands instead. REF_FORCE=1
overrides. The engine self-test (SNAP=./glm_tiny TF=1) is unaffected.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

2026-07-12 13:35:15 +02:00

..

scripts

Organize project tools and local workflows: c/tools, c/scripts, c/tests, root Makefile (flat C core untouched) (#22)

2026-07-10 10:08:39 +02:00

tests

Grammar-forced speculative drafts: GBNF grammar as a third draft source, guaranteed-accepted forced spans, lossless + opt-in (#48, #70)

2026-07-12 01:35:39 +02:00

tools

Windows 11 native port, phase 1: MinGW-w64 static build, compat shims, setup + docs (#40)

2026-07-11 12:59:49 +02:00

.gitignore

Windows 11 native port, phase 1: MinGW-w64 static build, compat shims, setup + docs (#40)

2026-07-11 12:59:49 +02:00

backend_cuda.cu

Tiered CUDA acceleration for routed experts (opt-in, CPU default untouched) + REPLAY fixture harness (#16)

2026-07-10 07:41:09 +02:00

backend_cuda.h

Tiered CUDA acceleration for routed experts (opt-in, CPU default untouched) + REPLAY fixture harness (#16)

2026-07-10 07:41:09 +02:00

coli

coli doctor: read-only setup/health diagnostics (path, config, shards, disk, RAM budget, placement) (#33)

2026-07-11 19:39:13 +02:00

compat.h

Windows 11 native port, phase 1: MinGW-w64 static build, compat shims, setup + docs (#40)

2026-07-11 12:59:49 +02:00

doctor.py

coli doctor: read-only setup/health diagnostics (path, config, shards, disk, RAM budget, placement) (#33)

2026-07-11 19:39:13 +02:00

glm.c

Guard against running the tiny oracle against a real model (#76)

2026-07-12 13:35:15 +02:00

grammar.h

Grammar-forced speculative drafts: GBNF grammar as a third draft source, guaranteed-accepted forced spans, lossless + opt-in (#48, #70)

2026-07-12 01:35:39 +02:00

iobench.c

iobench: Windows correctness — aligned free, 64-bit totals, full-range random offsets (also improves Linux offset coverage) (#64)

2026-07-12 01:19:40 +02:00

json.h

colibrì: pure-C GLM-5.2 (744B MoE) engine with disk-streamed experts

2026-07-05 20:57:25 +02:00

Makefile

Grammar-forced speculative drafts: GBNF grammar as a third draft source, guaranteed-accepted forced spans, lossless + opt-in (#48, #70)

2026-07-12 01:35:39 +02:00

olmoe.c

Windows 11 native port, phase 1: MinGW-w64 static build, compat shims, setup + docs (#40)

2026-07-11 12:59:49 +02:00

openai_server.py

Isolated sequence KV contexts: KVState extraction, KV_SLOTS serve contexts with per-slot persistence, budget-aware pool accounting (#29)

2026-07-10 16:19:43 +02:00

ref_glm.json

colibrì: pure-C GLM-5.2 (744B MoE) engine with disk-streamed experts

2026-07-05 20:57:25 +02:00

resource_plan.py

Windows 11 native port, phase 1: MinGW-w64 static build, compat shims, setup + docs (#40)

2026-07-11 12:59:49 +02:00

setup.sh

setup.sh: detect libomp by dylib presence, not brew prefix (false-positive fix by fabio-rovai, #47)

2026-07-11 13:02:34 +02:00

st.h

Windows 11 native port, phase 1: MinGW-w64 static build, compat shims, setup + docs (#40)

2026-07-11 12:59:49 +02:00

tier.h

REPIN follows live session heat (decaying heat map; .coli_usage stays the persistent signal) + disk→RAM→VRAM promotion (#26)

2026-07-10 12:48:53 +02:00

tok_unicode.h

Organize project tools and local workflows: c/tools, c/scripts, c/tests, root Makefile (flat C core untouched) (#22)

2026-07-10 10:08:39 +02:00

tok.h

colibrì: pure-C GLM-5.2 (744B MoE) engine with disk-streamed experts

2026-07-05 20:57:25 +02:00